ARTI-1

Building Corporate Associations: Consumer Attributions for Corporate Socially Responsible Programs

Pam Scholder Ellen Georgia State University

Deborah J. Webb University o f West Georgia

Lois A. Mohr Georgia State University

Corporate social responsibility (CSR) is often used as a key criterion in gauging corporate reputation. This re- search examined the influence of consumers' attributions on corporate outcomes in response to CSR. Researchers and managers have considered consumers' beliefs about CSR initiatives to be simplistic, serving either economic ends or reflecting sincere social concerns. The results o f two studies established that consumers'attributions were more complex than traditionally viewed, mirroring many of the motives ascribed to companies by managers and re- searchers. Rather than viewing corporate efforts along a self- or other-centered continuum, consumers differenti- ated four types of motives: self-centered motives that are strategic and egoistic and other-centered motives that are values driven and stakeholder driven. Consumers re- sponded most positively to CSR efforts they judged as val- ues driven and strategic while responding negatively to efforts perceived as stakeholder driven or egoistic. Attributions were shown to affect purchase intent as well as mediate the structure of an often

Keywords: corporate social responsibility; corporate associations; motives; altruism; cause mar- keting; corporate reputation; commitment; congruency

Journal of the Academy of Marketing Science. Volume 34, No. 2, pages 147-157. DOI: 10.1177/0092070305284976 Copyright �9 2006 by Academy of Marketing Science.

Aaker (2005) claimed that most industries are hostile or are becoming hostile, meaning that they are characterized by overcapacity, low margins, and intense competition. In such markets, creating, refining, or even repairing a com- pany's reputation with its stakeholders is key to success. To effectively compete, managers must remember that

the power o f a brand lies in what customers have learned, felt, seen, heard, and so forth about the brand as a result o f their experiences over time. In other words, the power of a brand is in what resides in the minds of customers. (Hoeffier and Keller 2002:79)

Brown and Dacin (1997:69) labeled "all the information about a company that a person holds" one's corporate as- sociations. The corporate associations held by an individ- ual "serve as the 'reality' o f the organization for that individual" (Brown, Dacin, Pratt, and Whetten 2006:105).

Corporate associations play an important role in corpo- rate outcomes, including reputation; corporate, product, and brand evaluations; purchase intent; and customer identification with a company (e.g., Brown and Dacin 1997; G u rh an -Can l i and B a t r a 2004; Li ch t enstein, Drumwright, and Braig 2004; Mohr and Webb 2005). The challenge is creating and managing corporate associations so that they evolve as central, enduring, and distinctive (Albert and Whetten 1985) links in the minds o f relevant stakeholders that result in a desired reputation. Many

148 JOURNAL OF THE ACADEMY OF MARKETING SCIENCE SPRING 2006

questions remain unanswered about the processes by which corporate behaviors and communications influence what stakeholders actually think about an organization (i.e., Viewpoint 4 of Brown et al. 2006) or how corporate associations influence corporate outcomes.

One type o f corporate association receiving attention in the literature and in practice is corporate social responsi- bility (CSR) associations. CSR associations are those that "reflect the organization's status and activities with respect to its perceived societal obligations" (Brown and Dacin 1997:68). Increasingly, CSR is being used as a key criterion in gauging corporate reputation (cf. the Harris- Fombrun Reputation Quotient, Fortune's 100 Best Com- panies to Work For). In pursuit of the opportunity to differ- entiate themselves from the competition and bolster their reputations, U.S. companies spent $9 billion in support of social causes in 2001 (Cone, Feldman, and DaSilva 2003). Despite these efforts, research indicates that CSR is a criti- cal area for improvement for companies. A Gallup poll found that confidence in big business is low, with only 7 percent of respondents saying they had a "great deal" and 17 percent saying that they had "quite a lot" of confidence (Roper Center at the University of Connecticut 2004). Forehand and Grier (2003:350) conceptualized this "consumer distrust or disbelief of marketer actions" as skepticism.

CSR takes many forms, including philanthropy, cause- related marketing, environmental responsibility, and humane employee treatment, among others. Regardless of their form, CSR efforts are generally intended to portray an image of a company as responsive to the needs of the society it depends on for survival.l Academic studies of CSR initiatives offer support for such strategies (Brown and Dacin 1997; Handelman and Arnold 1999; Lafferty and Goldsmith 1999; Molar and Webb 2005; Osterhus 1997). Research also suggests that critical intervening processes are important influences on whether the intended effects are achieved or whether CSR backfires (Forehand and Grier 2003; Handelman and Arnold 1999; Osterhus 1997). These intervening processes include the level of consumer trust in a firm (Lafferty and Goldsmith 1999; Osterhus 1997) or, more specifically, what motives consumers attribute to a firm's behavior. Evaluations o f a farm and its actions are considered to rest in part on the degree to which consumers associate egoistic (self- centered) or altruistic (other-centered) motives (Handel- man and Arnold 1999; Webb and Molar 1998). Thus, as Gilbert and Malone (1995) would predict, consumers may care less about what firms are doing than about why they are doing it.

This research sought to enhance our understanding of consumers' attributions about the motives behind CSR and how they influence corporate outcomes. In addition, we have begun the process of examining how elements of CSR initiatives influence consumers' attributions. Two

studies were conducted to examine the following ques- tions: (a) Are attributions elicited by CSR efforts simple bipolar judgments of altruistic or egoistic corporate motives, or are they more complex? (b) Are attributed motives pure or mixed; that is, are motives either self- or other centered, or are consumers capable of integrating both in judgments? (c) Do attributions vary with the nature of an offer? and (d) Do attributions mediate offer effects on relevant firm outcomes?

First, to avoid imposing a preconceived framework on consumers' attributions, an exploratory, qualitative study

�9 was undertaken to identify the range of motives consumers attribute to CSR efforts. The findings enabled us to explore the complexity and nature of motives elicited and their influence on firm evaluations. Then, the results of the qualitative study were used to offer a theoretical explana- tion for the attributions discovered and to develop a quanti- tative study. The quantitative study was used to develop and test a measure for assessing four types of consumers' attributions for CSR, to conduct an experiment to deter- mine whether attributions vary with the offer elements, and to measure the influence of resulting attributions on purchase intent. These issues were examined in the con- text of a common form of CSR, cause-related marketing.

ATTRIBUTIONAL INFERENCES ABOUT CSR

Drumwright (1996) found that even though managers described firms' motives as mixed--serving both eco- nomic as well as social objectives--these same managers believed that consumers are simplistic in their judgments about CSR initiatives and view them as either serving eco- nomic ends or reflecting sincere social concerns. Extant research on the impact of CSR efforts has taken a similar view, measuring motives along continuums such as "self- serving to society serving" and "firm serving to public serving" (cf. Barone, Miyazaki, and Taylor 2000; Fore- hand and Grier 2003; Lichtenstein et al. 2004). This is con- ceptually consistent with the minimum boundary condi- tion for any social actor of the distinction between the self versus the other (Whetten and Mackey 2002).

Yet there is research to suggest that responses to CSR may be more complex than represented in these approaches. Fein (1996:165) argued that suspicion of ulte- rior motives is likely to encourage individuals to "entertain multiple, plausible rival hypotheses about the motives or genuineness" of an entity's behavior. In a climate of lim- ited trust, all deeds may be heavily scrutinized, yielding more complex assessments of motives. When expecta- tions are disconfirmed, people "give much thought to 'why' questions" (p. 165), leading to more sophisticated attributions. Since consumers show little confidence and trust in business, CSR efforts to appear as a "good citizen" might promote such attention. While many suggest that

Ellen et al. / BUILDING CORPORATE ASSOCIATIONS 149

inconsistency or duality is difficult for consumers to rec- oncile, Williams and Aaker (2002) argued that consumers are capable, when presented with persuasive communica- tions, of accepting and synthesizing apparently contradic- tory information in making judgments rather than relying on more simplistic bipolar views. They found that positive and negative emotional reactions co-occur when individu- als are exposed to ads with mixed emotional appeals.

STUDY 1

To discover the range of motives that might be elicited in response to cause-related activities, open-ended ques- tions were used. Students (n = 281) at a major university participated in a study of customer opinions of business practices. Respondents ranged in age from 19 to 52 years (M = 26.3 years), and most were employed (44% full-time, 41% part-time). The sample was gender balanced and eth- nically diverse (63% Caucasian, 21% African American, and 16% other).

To generate maximally different attributions, respon- dents were exposed to hypothetical radio scripts request- ing consumers' charitable donations through different types of stores for different types of causes. Each person saw one cause-related marketing offer and was asked to record any thoughts about why the company would make such an offer and their overall evaluation of the fkrm on a four-item attitude scale. Responses were first divided into discrete thoughts by two i n d e p e n d e n t coders, who resolved disagreements through discussion. The result was 647 relevant thoughts for why the company made the offer. With a range of 0 to 12 thoughts per person, the average number of attributions was 2.28.

Following Miles and Huberman (1994), descriptive codes were developed by combining similar reasons to form a smaller number of categories based directly on the data. Using four independent coders, the coding system was developed using an iterative process of coding a sam- ple of thoughts, discussing disagreements, modifying the coding system, coding another sample of thoughts, and so on. The result was 17 categories of reasons for why com- panies would make the offer (see Table 1). Two coders conducted the final coding, resolving disagreements in discussions of the entire team. The coders agreed on their categorization of 87 percent of the thoughts. The propor- tional reduction in loss reliability (Rust and Cooil 1994) was .92.

Results

The elicited attributions showed a complex range o f motives, demonstrating that at least some consumers rec- ognized a variety of influences on a company's decision to engage in CSR. At one end of the range, the attributions

indicated that some consumers saw companies as truly caring or as getting involved because the companies believed that they were morally obligated or at least expected to help. At the opposite end were the most extreme, self-centered motives, such as tax write-offs or a suggested "pocketing" o f donations. In between were a variety o f expected business practices, such as building customer loyalty, getting more customers and sales, and improving a company's image. To gain a perspective on the dominant motives, the attributions were grouped into three more inferential and explanatory categories: other centered, self-centered, and win-win. See Table 1 for descriptive statistics.

While self-centered motives were the most frequently mentioned, most respondents (74%) gave more than one attribution. To examine whether respondents attributed pure or mixed motives, a ratio o f self-centered to total thoughts was created by dividing each person's number of self-centered motives by his or her total number o f thoughts. The result was a score ranging from 0 to 1, where 0 represented no self-centered thoughts and 1 represented only self-centered attributions. While 42.8 percent attrib- uted pure motives to the company (all either self-centered or other-centered), 57.2 percent saw the co mpany's motives as mixed, with 26.7 percent making equal num- bers o f self- and other-centered attributions. Interestingly, when attributions were mixed, the evaluation of the fn-rn was more positive (M = 6.05 on a 7-point scale) than when attributions were purely self-centered (M = 5.43) or purely other centered (M = 5.50) (F = 7.84, p < .01).

Discussion

The motives elicited fro m consumers in Study 1 showed commonality with those proposed by Swanson (1995). In her reorientation o f the corporate social perfor- mance model, Swanson bridged management and busi- ness ethics research to propose three principal motivations for companies to engage in CSR: economic, positive duty, and negative duty. Economic motives, usually the focus of management researchers, incorporate a firm's perfor- mance objectives, such as sales, profit, and return on investment. The duty-aligned perspectives are usually adopted by ethical researchers and focus on corporate moral behaviors and the associated obligations to society. Positive duty recognizes that a company may be involved in CSR to help others, while negative duty holds that a company's motivation may be an exercise in restraint to meet stakeholder expectations. Similar motives were iden- tified by Maignan and Ralston (2002) in their review of companies' self-presentations on their Web pages, renam- ing them performance driven, value driven, and stake- holder driven, respectively.

While all three motives are legitimate from corpora- tions' perspectives, as components of their reputations

150 JOURNAL OF THE ACADEMY OF MARKETING SCIENCE SPRING 2006

TABLE 1 Open-Ended Attributions About Company Motives

Attribution Frequency % of Responses % of Cases

Other centered 232 34.5 63.5 They care/want to help 138 21.3 Identify with victims/beneficiaries 23 3.6 Owe the community 16 2.5 Morally obligated (internal) 16 2.5 Company has what's needed/little cost to help 14 2.2 Help customers to help 12 1.9 Owe the community/expected to help 6 0.9

Self-centered 414 63.7 93.7 Affect what people think about them 231 32.9 Get more customers/sales 158 24.4 Tax write-off 21 3.2 Help themselves (general) 8 1.2 Build customer loyalty 5 0.8 Helping so company can survive 4 0.6 Pocket the donations 3 0.5 Competitive advantage 2 0.3 Compensate for previous bad deeds I 0.2

Win-win 7 1.1 2.7

with consumers, it is important to determine whether they represent a simple continuum from self-serving to other serving: performance driven --~ stakeholder driven --~ val- ues driven. I f so, then firms would need to manage their messages to emphasize the values-driven aspects and de- e m p h a s i z e the p e r f o r m a n c e - d r i v e n aspects to these publics.

Two findings o f Study I suggest a different model. First, those consumers who attributed both other-centered and self-centered motives reported more positive respons- es to the firm than those who attributed either one or the other. Rather than self-centered motives being viewed as negative, c o n s u m e r s ' performance-driven motives seemed to fall into typical strategic goals (e.g., getting more cus- tomers and sales) and highly egoistic motives (e.g., pock- eting the donations). As Whetten and Mackey (2002) sug- gested, attributions related to typical strategic goals o f getting and keeping customers are inherent in the exis- tence o f a firm as a social actor and are widely accepted. However, attributions such as taking advantage o f a cause or nonprofit have negative, egoistic connotations and are not likely to be widely accepted.

S e c o n d , the d u t y - a l i g n e d goals w e r e defined b y Swanson (1995) as positive when they were designed to help others and negative when they were dictated b y stake- holders. This suggests that consumers will evaluate CSR efforts more positively when they are driven apparently by corporate values and more negatively when they are in re- sponse to stakeholder requirements.

Hypothesis 1: Purchase intent in response to CSR associ- ations will be higher when attributions are (a) values

driven or (b) strategic and lower when attributions are (c) egoistic or (d) stakeholder driven.

Study 2 was conducted to develop and test a measure to assess consumers' attributed motives for C S R and to deter- mine whether these attributions were elicited in response to a cause-related marketing offer. Furthermore, we exam- ined whether attributions were differentially affected b y the elements o f the offer. Finally, we examined whether the resulting attributions mediated the influence o f the offer elements on purchase intent.

STUDY 2

To examine whether attributions play a mediational role in consumers' processing o f CSR initiatives, the manipulated elements o f the cause-related marketing offer had to have a significant influence on corporate outcomes. Thus, two offer elements were selected on the basis o f their wide acceptance as influential on c o n s u m e r s ' responses to CSR: (a) the fit o f the cause's mission with the c o m p a n y ' s core business and (b) the level o f c o m p a n y c o m m i t m e n t to the cause.

Company-Cause Fit

Early corporate donors selected the causes "least asso- ciated with their line o f business" (Smith 1994:107), fear- ing opportunistic attributions. Drumwright (1996) found that high fit between a company and a cause led managers to fear cynical reactions from consumers, who might view

Ellen et al. / BUILDING CORPORATE ASSOCIATIONS 151

the company as exploiting the cause. In an experiment, Ellen, Mohr, and Webb (2000) found that offers judged as less congruent were evaluated marginally more positively than congruent offers. However, researchers and practitio- ners consistently r e c o m m e n d that companies support causes that are logically matched to their product lines,

b r a n d images or positioning, or target markets (cf. Cone et al. 2003; Varadarajan and Menon 1988). A close match between a company's core business and a cause is likely to lead consumers to perceive the company as more expert and transfer more positive feelings about the cause to the company (Hoeffler and Keller 2002). Becker-Olsen, Cudmore, and Hill (2006) found that low-fit CSR initia- tives had a negative influence on consumers' beliefs, atti- tudes, and purchase intent. Fein's (1996) work suggests that a close match is less likely to raise suspicion because a firm is not acting out of character with its prime directive; thus, attributions to the firm as strategic would be ex- pected. Similarly, when suspicions are not raised, values- driven attributions are more likely, while egoistic and stakeholder attributions are less likely.

Hypothesis 2: High (low) fit between a cause's mission and a company's core business will increase (de- crease) (a) values-driven and (b) strategic attribu- tions while decreasing (increasing) (c) egoistic and (d) stakeholder-driven attributions.

C o m m i t m e n t to a C a u s e

L'Etang (1994) argued that the commitment of a com- pany to a cause is a major factor determining whether the company is seen as exploiting the cause. Dwyer, Schurr, and Oh (1987) defined commitment as "an implicit or ex- plicit pledge of relational continuity between exchange partners" (p. 19). They described three factors that lead to perceived commitment: the amount of input, the durability of the association, and the consistency (stability) o f input. Webb and Mohr (1998) found that the length of time com- mitted to a cause was used as a cue for judging a firm's mo- tives: longer term commitments were viewed as more well intentioned, while shorter term campaigns were viewed as just another way to increase sales. Drumwright (1996) found that employees judged social advertising campaigns as more successful when the campaigns extended over multiple years, while those lasting 6 months or less "in- variably spelled doom" (p. 81). Varadarajan and Menon (1988) suggested that a medium- or long-term commit- ment provides more time for consumers to learn about a company-cause connection, and hence, there is more time for public relations to become effective. A longer commit- ment is likely to indicate a "real" commitment to the effort, thus suggesting values-driven motives. A shorter commit- ment might be viewed as reactive, driven by strategic

performance demands, egoistic motives, or stakeholder pressure.

Hypothesis 3: High (low) commitment to a cause will in- crease (decrease) (a) values-driven attributions while decreasing (increasing) (b) strategic, (c) ego- istic, and (d) stakeholder-driven attributions.

As discussed above, prior research has found that inter- vening processes take place in consumers' evaluative pro- cessing o f CSR initiatives (Forehand and Grier 2003; Handelman and Arnold 1999; Lafferty and Goldsmith 1999; Osterhus 1997; Webb and Mohr 1998). We believe that attributions are among these intervening processes.

Hypothesis 4: Attributions will mediate the relationship between offer elements and purchase intent.

Method

To determine whether fit and commitment differen- tially affected attributions, fictitious ads for a cause- related marketing offer were created to yield a 3 (fit: high, no relationship, low) x 2 (commitment: high, low) between-subjects experiment with a control-group (no cause-related offer) design. A survey was mailed to a ran- dom sample of 490 staff employees at a large university. A follow-up reminder notice was sent 1 week later, yielding 193 usable surveys, for a 44.7 percent response rate. Each participant was asked to read a scenario and shown a ran- domly assigned fictitious ad from a "new campaign run- ning in major newspapers around the United States"

Experimental manipulations. To minimize product or quality differences, a pretest indicated that "The Gas Sta- tion" met the criteria of providing a frequently purchased necessity by most consumers and operating in a parity market. In a second pretest, 3 causes were selected from 15 causes described as important to most people in The Gal- lup Poll Monthly (1996). "The Gas Station" was perceived to fit best with a cause that provided "transportation for older and disabled members of our community. They pro- vide transportation for these individuals to get to places such as the doctor's office, drug store, e t c " (M = 6.16). Low fit was one seeking "to protect wildlife habitats. They fight against the building o f roads and highways that re- quires clear-cutting of local forests" (M = 4.43). The cause selected as having no obvious relationship with the firm's business was "fights homelessness. They renovate empty buildings into affordable apartments for homeless families in our community" (M = 4.98).

Using Drumwright's (1996) guidelines, commitment was manipulated at two levels (high and low) by varying the length of time the cause-related marketing offer was in effect. High commitment was described as providing sup- port to the cause "as it has for the past 7 years," while low

152 JOURNAL OF THE ACADEMY OF MARKETING SCIENCE SPRING 2006

commitment was described as lasting for 1 month (e.g., "April"). The contribution to the cause was held constant at 1 percent across treatments.

Scenario. All participants were asked to imagine that recently, "The Gas Station" had opened a location that was as convenient for them as their current stations and offered the same quality, price, and service. An ad made claims about high quality; low prices; and modem, convenient lo- cations. A cause-related marketing offer was also made in the six treatment ads. Participants were asked how likely they would be to switch, using four 7-point, semantic dif- ferential items (e.g., unlikely and likely; Oliver and Swan 1989; ~ = .94). Next, they responded to a 21-item, 7-point, Likert-type scale developed to tap perceived motives on the basis of the attributions revealed in Study 1. Finally, manipulation checks and demographic questions were completed.

Results

Respondents were 60 percent female and had an aver- age age of 38.2 years. Household incomes ranged from under $15,000 to over $250,000, with 44 percent between $25,000 and $44,999 and 24 percent between $45,000 and $74,999. Race was diverse (54% Caucasian, 38% African American).

Common-factor analysis with oblimin rotation was used to examine the structure o f the 21-item attribution scale, because Study 1 indicated that there would be corre- lations among some of the attributions. Four factors in- cluding 16 items, accounting for 62.8 percent of the vari- ance, were extracted (see Table 2) The first factor, labeled values-driven attributions (variance extracted 26.6%), in- cluded motives such as caring about the cause. The second factor, labeled stakeholder-driven attributions (variance extracted 16.6%), reflected a response to the expectations o f different stakeholders. The third factor, egoistic attribu- tions (variance extracted 11.2%), was composed o f 4 items attributing the firm's participation to more blatant self- centered reasons (e.g., taking advantage of the cause). The fourth factor, strategic attributions (variance extracted 8.4% ), consisted o f 3 items that attributed the firm's partic- ipation to self-centered goals reflecting typical business objectives (e.g., making a profit).

Scales for each attribution factor were created (see Table 2 for ~ values and means). The individual items for each factor were weighted using factor scores and sum- med. These were used to determine whether attributions were affected differentially by the elements of the offer and whether these in turn mediated the influence o f the offer on purchase intent.

Manipulation checks. To assess fit, three 7-point, Likert-type items adapted from Sengupta, Goodstein, and B oninger (1997) assessed the fit, relevance, and appropri-

ateness o f the partnership between the ftrrn and the cause (~ = .94). Significant differences were found between the means o f the manipulation check across the three manipu- lated levels o f fit, F(2, 159) = 22.93, p < .01. However, ex- amination o f the means indicated that only two signifi- cantly different levels o f fit were created. The high-fit group (transportation) was significantly higher (M = 5.25) than the other two groups, which did not differ signifi- candy from each other (homelessness M = 3.60, wildlife M = 3.79). These two treatments were collapsed.

Commitment was assessed with two 7-point semantic differential items (e.g., the fn-m was committed to and cared about the cause; ~ = .86). Two significantly different levels of firm commitment were attained, F(1, 155) = 5.77, p < .02. Those receiving the low-commitment treatment (i.e., "April") rated commitment significantly lower (M = 3.91) than those who received the high-commitment (i.e., 7-year) treatment (M = 4.42).

The effects o f fit and commitment on values-driven, stakeholder-driven, egoistic, and strategic attributions were tested using multivariate analysis o f variance (MANOVA). While there were no significant interactions, the influence of fit on attributions was significant, Wilks's

= .90, F(4, 143) = 4.22, p < .01, ~2 =. 11, as was commit- ment, Wilks's ~, = .94, F(4, 143) = 2.19, p < .07,1] ~ = .06. Participants exposed to the high-fit treatments were more likely than those in the low-fit treatments to attribute ftrrn participation in the cause-related marketing offer to values-driven and strategic reasons, values-driven F(1, 149) = 6.41,p < .01, ~2 =.04; strategic F(1,149) = 6.41,p < .01, lq 2 = .04, and less likely to attribute the offer to egoistic reasons, F ( 1 , 1 4 9 ) = 5.17, p < .02, 112 = .03, offering sup- port for Hypotheses 2a, 2b, and 2c, respectively. However, Hypothesis 2d was not supported: fit did not have a signifi- cant influence on stakeholder-driven attributions.

While it did not influence values-driven, egoistic, or strategic attributions, commitment had a significant influ- ence on stakeholder-driven attributions, F ( 1 , 1 4 9 ) = 3.46, p < .07, rl 2 = .02. Lo w commitment led to higher stake- holder-driven attributions than high commitment. Thus, Hypotheses 3a to 3c were not supported, but Hypothesis 3d was supported.

Next, following Baron and Kenny's (1986) proce- dure, regression was used to determine i f the treatments affected purchase intent. Fit had a significant influence on purchase intent, F ( 1 , 1 6 2 ) = 4.40, p < .04, adjusted R 2 = .02, b = .58, t = 2.10, p < .04. Surprisingly, commitment did not. Only the mediational effects for fit were examined subsequently.

Regression was used to regress the four attributions scales on purchase intent, yielding a significant model, F(4, 149) = 10.34, p < .01, adjusted R 2 = .20. Each of the four attributions had a significant influence on purchase intent. Values-driven and strategic attributions increased purchase intent (values driven b = .10, t = 4.00, p < .01;

r

0

c t~

I::: o

o

, . o

E , . i

0 o

t o

k .

I I I

. ~ ' ~

u ~ u~

.'-' = = m

~ ' ~

4 ~ ' ~ &

7, .~

e ~

V

e .

o ,-.1

V

o ' , t--i

II

6 6 t'-:.

II

co

5

z

1 5 3

154 JOURNAL OF THE ACADEMY OF MARKETING SCIENCE SPRING 2006

strategic b = . 13, t = 2.42, p < .02), supporting Hypotheses l a and lb. Also, as predicted, egoistic and stakeholder- driven attributions decreased purchase intent (egoistic b = -.11, t = -2.95, p < .01; stakeholder driven b = -.06, t = -1.94, p < .05), supporting Hypotheses lc and ld.

A series of regressions determined whether attributions mediated the influence of fit on purchase intent. First, three regressions reconf'n-med that higher fit had a signifi- cant positive influence on values-driven, F(1,157 ) = 6.20, p < .01, adjusted R 2 = .03, b = 2.16, t = 2.49, p < .01, and strategic, F ( 1 , 1 5 6 ) = 5.68, p < .02, adjusted R 2 = .03, b = .95, t = 2.38, p < .02, attributions and a negative influence on egoistic attributions, F(1,158) = 5.51,p < .02, adjusted R 2 = .03, b = -1.34, t =-2.35, p < .02. As in the MANOVA, fit did not have a significant influence on stakeholder- driven attributions.

Then, a regression including values-driven, egoistic, and strategic attributions and fit established the media- tional role o f these attributions, F(4, 152) = 10.49,p < .01, adjusted R 2 = .20, supporting Hypothesis 4. While the attributions had a significant influence on purchase intent (values-driven b = .09, t = 3.74, p < .01; egoistic b = - . 11, t = -2.93, p < .01; strategic b = .10, t = 1.83, p < .07), the influence of fit on purchase intent was no longer signifi- cant (b = .29, t = 1.04, p < .30) when attributions were included in the regression.

Overall, these results indicate that fit affected respon- dents' purchase intent through its influence on attribu- tions. When there was a high level o f fit between the com- pany's business and the cause, the company was seen as getting involved because of its desire to help the cause and to build relationships with customers rather than for exces- sive profiteering. This then led to a greater willingness to switch than when the cause did not fit with the firm's business.

While fit influenced values-driven, strategic, and egois- tic attributions, and they subsequently influenced switch- ing, no such mediational role was found for stakeholder- driven attributions. Stakeholder-driven attributions were affected only by commitment, and commitment did not have a significant influence on purchase intent.

DISCUSSION

This research examines the attributions made by con- sumers about the motives underlying companies' partici- pation in CSR. The results of two studies establish that consumers' attributions play an important role in their responses to CSR. Additionally, attributions are more complex than traditionally viewed, mirroring many o f the motives ascribed to the companies by managers and researchers (Drumwright 1996; Swanson 1995). The range o f motives identified in both studies showed that self- and other-centered motives were further differenti-

ated into components that differed in valence. Specifically, consumers distinguished between self-centered motives that were strategic and egoistic, reacting positively and negatively, respectively, to these motives. Likewise, other- centered motives were differentiated, with values-driven motives viewed positively and stakeholder-driven motives perceived negatively. Prior research has seemed to pre- sume that consumer responses to CSR efforts are linearly related to the degree to which programs are viewed as self- or other centered, Instead, we find that the majority o f par- ticipants attributed company participation to a combina- tion o f both. Response to the firm and its offer was most positive when consumers attributed both values-driven 'and strategic motives to the firm.

To establish that attributions varied by the nature of the offer and were thus controllable, we examined the influ- ence o f fit between the company's core business and the selected cause and length o f commitment. While higher fit makes it easier to see why a company benefits by engaging in the effort, it also might raise the specter o f opportunism. On the basis o f Fein's (1996) work, higher fit apparently reduced the suspicion, allowing consumers to attribute "typical" business motives (i.e., strategic) as well as values- driven motives. Lower fit, which might raise suspicions, led to more egoistic attributions. Furthermore, values- driven, egoistic, and strategic attributions mediated the relationship between fit and purchase intent. Apparently, high-fit matches between the company's business and the cause led consumers to believe that the company was moti- vated by a desire to help others in the normal conduct of its business affairs rather than a desire to selfishly use the cause. It is these attributions that led to higher purchase intent.

As with higher fit, longer time commitments were expected to lead to more values-driven attributions, while shorter commitments might be viewed as more reactive to pressures o f stakeholders or business demands. When commitment was only for a short period o f time, partici- pants thought that the company was participating only to meet others' expectations instead o f acting on the princi- ples o f the organization. Furthermore, stakeholder-driven attributions decreased purchase intent. It appears that con- sumers do not give credit to companies that engage in CSR because o f pressure from customers and other stake- holders. This is conceptually consistent with the negative- duty perspective prevalent in business ethics research (Swanson 1995).

IMPLICATIONS AND FUTURE RESEARCH

Our research offers insights into Brown et al.'s (2006) Viewpoint 4, addressing what stakeholders actually think o f an organization. Attributions were found to mediate the relationship between the elements o f CSR offers and

Ellen et al. / BUILDING CORPORATE ASSOCIATIONS 155

consumers' responses to the firm. The findings provide strong support for measuring consumer attributions about the motives behind companies' participation in CSR initia- tives. Furthermore, the structure of CSR initiatives influ- ences the type of attributions that are made in response to CSR, and these attributions affect corporate outcomes. Much previous research has focused on global evaluations and intent; this research suggests that these responses are likely to be mediated by attributions of firms' motives for CSR. The multidimensional measure developed and vali- dated to assess consumers' attributions can be used in future research to examine their role in determining these more global responses to CSR.

The attributions elicited in this research proved more dimensional than typically examined. Rather than simple unidimensional attributions (e.g., self- vs. other centered), four different types of attributions with different effects were identified. In addition, it was found that consumers evidently dealt with the duality of other- and self-centered motives and in fact responded more positively when both existed. Thus, while consumers and the public may look cynically at businesses, they recognize and apparently expect that businesses can serve two masters: their bottom lines and long-run viability and the needs o f society. These findings support the work done by Williams and Aaker (2002) concerning the acceptance of duality in persuasive communications. They found that positive and negative emotional reactions c o - o c c u r when individuals are exposed to ads with mixed emotional appeals. Likewise, we establish that individuals exposed to a cause-related marketing offer attribute company participation to a com- bination of self- and other-centered motives. Furthermore, they distinguish between positive and negative self- and other-centered motives. Just as Swanson (1995) would predict, consumers are able to reconcile the self- and other-centered motives o f strategic and values-driven motives, with both having a positive influence on purchase intent. It is important to note that this means that managers do not have to hidestrategic aspects of CSR. These find- ings also offer support for Forehand and Grier's (2003) work indicating that the negative influence of consumer skepticism can be inhibited by acknowledging the strate- gic benefits to a firm. Thus, caution must be exercised in developing and implementing CSR programs. Future research is needed to understand how managers can com- municate the dual motives of CSR programs to achieve favorable values-driven and strategic consumer attribu- tions while avoiding egoistic and stakeholder-driven attributions.

The lack of a significant relationship between commit- ment and participant responses may be because only one dimension of commitment identified by Dwyer et al. (1987), the durability of the relationship (i.e., the length of time the retailer supported the cause) was manipulated. The other two dimensions, the amount of input and the

consistency o f input, were held constant. Consumers may examine the three dimensions holistically rather than as individual pieces o f information. Further research is need- ed to identify the elements o f CSR that signal commitment to the consumer.

Future research is needed to evaluate the generaliz- ability o f the findings. Similarly, the use of a hypothetical product removes the influence o f preexisting beliefs about the motives o f specific industries on the basis of cus- tomer experience. The research o f Hilton, Fein, and Miller (1993) suggested that when individuals suspect that an actor may purposefully appear to disconfirm negative expectancies, many of the actions that might normally result in expectancy disconfirmation lose their potency. If consumers have preexisting beliefs that firms intention- ally engage in CSR to make up for their shortcomings, CSR may lose its viability as a marketing tool. Research is needed to determine the extent to which such stigmas exist.

Our findings, coupled with those of Barone et al. (2000), indicate that understanding consumers' attribu- tions of a farm's motivation for engaging in CSR becomes even more important in nonparity markets. Barone et al. found that consumers engage in compensatory processing when presented cause-related marketing offers in non- parity markets. This means that they make trade-offs between products sold with cause-related marketing offers and lower priced or higher quality products of com- petitors. Thus, when managers design offers in nonparity markets, not only must they be concerned with consumers' attributions of self-centered behavior, they must also be concemed with the strength o f the attributions as the cus- t o m e r weighs t h ei r o ffers against the alternatives. Future research is needed to understand how trade-offs are made between CSR associations and other corporate associations.

These findings highlight the importance o f designing and implementing CSR initiatives at the strategic level along with other important aspects of firm positioning, such as price and quality. Otherwise, consumers may per- ceive such efforts as tactical maneuvers designed to increase sales by misleading customers. Handelman and Arnold's (1999) findings suggest that consumers have a minimally acceptable level o f CSR for firms within a given field. Research is needed to understand the role attribu- tions play in determining this minimally acceptable level of CSR and whether being the first to exceed it can provide a company with a first-mover advantage. Further research is also needed to understand how CSR associations evolve as central, enduring, and distinctive to a firm's reputation (Albert and Whetten 1985).

The level of corporate investment in social causes makes it clear that CSR is viewed as key for many firms to build reputation and create differential advantage. This research indicates that consumer reaction to such initia-

156 JOURNAL OF THE ACADEMY OF MARKETING SCIENCE SPRING 2006

tives is in fact complex and incorporates duality of motives. Future research will be needed to deepen our understanding of consumers' attributional analyses in response to CSR associations and their influence on corporate outcomes.

ACKNOWLEDGMENTS

This research was sponsored in part by grants from the Georgia State University Dissertation Grant Program and the Robinson College of Business.

NOTE

1. Other authors have presented extensive discussions on what consti- tutes CSR. Those issues are not addressed here; the reader is referred to articles such as that by Smith (2003).

REFERENCES

Aaker, David A . 2005. Strategic Market Management (7th ed.). New York: John Wiley.

Albert, Stuart and David A. Whetten. 1985. "Organizational Identity?' In Research in Organizational Behavior. Eds. L. L. Cummings and B. M. Straw. Greenwich, CT: JAI, 263-295.

Baron, Reuben M. and David A. Kenny. 1986. "The Moderator-Mediator Variable Distinction in Social Psychological Research: Conceptual, Strategic, and Statistical Considerations?' Journal of Personality and Social Psychology 51 (6): 1173-1182.

Barone, Michael J., Anthony D. Miyazaki, and Kimberly A. Taylor. 2000. "The Influence of Cause-Related Marketing on Consumer Choice: Does One Good Turn Deserve Another?" Journal of the Academy of Marketing Science 28 (2): 248-262.

Becker-Olsen, Karen L., B. Andrew Cudmore, and Ronald Paul Hill. 2006. "The Impact of Perceived Corporate Social Responsibility on Consumer Behavior." Journal of Business Research 59 (1): 46-53.

Brown, Tom J. and Peter A. Daein. 1997. "The Company and the Product: Corporate Associations and Consumer Product Responses?' Journal of Marketing 61 (January): 68-84.

, Michael G. Pratt, and David A. Whetten. 2006. "Iden- tity, Intended Image, Construed Image, and Reputation: An Interdis- ciplinary Framework and Suggested Terminology." Journal of the Academy of Marketing Science 34(2): 99-106.

Cone, Carol L., Mark A. Feldman, and Alison T. DaSilva. 2003. "Causes and Effects." Harvard Business Review 81 (July): 95-101.

Drnmwright, Minette E. 1996. "Company Advertising With a Social Di- mension: The Role of Noneconornic Criteria." Journal of Marketing 60 (October): 71-87.

Dwyer, E Robert, Paul H. Schurr, and Sejo Oh. 1987. "Developing Buyer-Seller Relationships." JournalofMarketing 51 (April): 11-27.

Ellen, Pare Scholder, Lois A. Mohr, and Deborah J. Webb. 2000. "Chari- table Programs and the Retailer: Do They Mix?" Journal of Retailing 76 (3): 393-406.

Fein, Steven. 1996. "Effects of Suspicion on Attributional Thinking and the Correspondence Bias." Journal of Personality and Social Psy- chology 70 (6): 1164-1184.

Forehand, Mark R. and Sonya Grier. 2003. "When Is Honesty the Best Policy? The Effect of Stated Company Intent on Consumer Skepti- cism." Journal of Consumer Psychology 13 (3): 349-356.

The Gallup Poll Monthly. 1996. 364 (January): 34-35. Gilbert, Daniel T. and Patrick S. Malone. 1995. "The Correspondence

Bias." Psychological Bulletin 117 (1): 21-38. Gurhan-Canli, Zeynep and Rajeev Batra. 2004. "When Corporate Image

Affects Product Evaluations: The Moderating Role of Perceived Risks." Journal of Marketing Research 41 (May): 197-205.

Handelman, Jay M. and Stephen J. Arnold. 1999. "The Role of Marketing Actions With a Social Dimension: Appeals to the Institutional Envi- ronment?' Journal of Marketing 63 (July): 33-48.

Hilton, James L., Steven Fein, and Dale T. Miller. 1993. "Suspicion and Dispositional Inference." Personality and Social Psychology Bulletin 19 (October): 501-512.

Hoeffler, Steve and Kevin Lane Keller. 2002. "Building Brand Equity Through Corporate Societal Marketing." Journal of Public Policy and Marketing 21 (1): 78-89.

Lafferty, Barbara A. and Ronald E. Goldsmith. 1999. "Corporate Credi- bility's Role in Consumers' Attitudes and Purchase Intentions When a High Versus Low Credibility Endorser is Used in the Ad." Journal of Business Research 44 (February): 109-116.

L'Etang, Jacquie. 1994. "Public Relations and Corporate Social Respon- sibiIity: Some Issues Afising?'JournalofBusiness Ethics 13 (2): 111- 123.

Lichtenstein, Donald R., Minette E. Drumwright, and Bridgette M. Braig. 2004. "The Effect of Corporate Social Responsibility on Cus- tomer Donations to Corporate-Supported Nonprofits." Journal of Marketing 68 (October): 16-32.

Maignan, Isabelle and David A. Ralston. 2002. "Corporate Social Re- sponsibility in Europe and the U.S.: Insights From Businesses' Self- Presentation?' Journal of International Business Studies 33 (3): 497- 514.

Miles, Matthew B. and A. Michael Huberman. 1994. Qualitative Data Analysis. Thousand Oaks, CA: Sage.

Mohr, Lois A. and Deborah J. Webb. 2005. "The Effect of Corporate So- cial Responsibility and Price on Consumer Responses?' Journal of Consumer Affairs 39 (1): 121-147.

Oliver, Richard L. and John E. Swan. t 989. "Consumer Perceptions of In- terpersonal Equity and Satisfaction in Transactions: A Field Survey Approach?' Journal of Marketing 53 (April): 21-35.

Osterhus, Thomas L. 1997. "'Pro-Social Consumer Influence Strategies: When and How Do They Work?" Journal of Marketing 61 (October): 16-29.

Roper Center at the University of Connecticut. 2004. Public Opinion On- line, Conducted by Gallup Organization, May 31, 2004. Retrieved f r o m h t t p : / / w e b . l e x i s - n e x i s . c o m / u n i v e r s e / d o c u m e n t ? _ m = 4e202c256342110260b891 d0795578e6& d o c n u m = 2 & w c h p = dGLbVzb-zSkVb&_md5=34e 1 bb3d3e600c4f85ee56b392decd46

Rust, Roland T. and Bruce Cooil. 1994. "Reliability Measures for Quali- tative Data: Theory and Implications?' Journal of Marketing Re- search 31 (February): 1-14.

Sengupta, Jaideep, Ronald C. Goodstein, and David S. Boninger. 1997. "All Cues Are Not Created Equal: Obtaining Attitude Persistence Un- der Low-Involvement Conditions?' Journal of Consumer Research 23 (March): 351-361.

Smith, Craig. 1994. "The New Corporate Philanthropy." Harvard Busi- ness Review 72 (May/June): 105-116.

Smith, N. Craig. 2003. "Corporate Social Responsibility: Whether or How?" California Management Review 45 (June): 52-76.

Swanson, Diane L. 1995. "Addressing a Theoretical Problem by Reori- enting the Corporate Social Performance Model." Academy of Man- agement Review 20 (1): 43-64.

Varadaraj an, P. Rajan and Anil Menon. 1988. "Cause-Related Marketing: A Coalignment of Marketing Strategy and Corporate Philanthropy." Journal of Marketing 52 (July): 58-74.

Webb, Deborah J. and Lois A. Mohr. 1998. "A Typology of Consumer Re- sponses to Cause-Related Marketing: From Skeptics to Socially Con- cerned?' Journal of Public Policy and Marketing 17 (Fall): 226-238.

Whetten, David A. and Alison Mackey. 2002. "A Social Actor Concep- tion of Organizational Identity and Its Implications for the Study of Organizational Reputation?' Business & Society 41(4): 393-414.

Williams, Patti and Jennifer L. Aaker. 2002. "Can Mixed Emotions Peacefully Coexist?" Journal of Consumer Research 28 (March): 636-649.

ABOUT THE AUTHORS

P a m S c h o l d e r E l l e n ( p e l l e n @ g s u . e d u ) is a n associate p r o f e s s o r o f m a r k e t i n g at the R o b i n s o n C o l l e g e o f B u s i n e s s at G e o r g i a

Ellen et al. / BUILDING CORPORATE ASSOCIATIONS 157

State University. Her research focuses on perceptual biases in the way consumers respond to marketplace offers, particularly in the public policy arena. In addition to the Journal of the Academy of Marketing Science, her research has been published in the Jour- nal of Consumer Research, the Journal of Public Policy & Mar- keting, the Journal of Consumer Affairs, and the Journal of Retailing.

D e b o r a h J. W e b b (dwebb@westga.edu) is an assistant profes- sor of marketing at the Richards College of Business at the Uni- versity of West Georgia. Her research interests include consum- ers' responses to prosocial corporate behaviors, marketing and society issues, and donation behavior. Her research has been pub-

lished in the Journal of the Academy of Marketing Science, the Journal of Retailing, the Journal of Public Policy & Marketing, the Journal of Consumer Affairs, and Marketing Education Re- view, among others.

Lois A. M o h r is a retired associate professor of marketing from the Robinson College of Business at Georgia State University. Her research interests focus on consumer responses to corporate social responsibility and services marketing. She has published in the Journal of Marketing, the Journal of Retailing, the Journal of Business Research, the Journal of Public Policy & Marketing, the Journal of Consumer Affairs, and the Journal of Services Marketing.